Growth, yield and nitrogen allocation in two rice cultivars under field conditions in venezuela

Alejandro J Pieters, Susana El Souki and César Nazar

Alejandro J Pieters. Biologist, Universidad Central de Venezuela (UCV). Ph.D. in Biological Science, University of Nottingham, UK. Researcher, Instituto Venezolano de Investigaciones Científicas (IVIC), Venezuela. Address: Centro de Ecología, IVIC. Carretera Panamericana Km 11. Caracas 1020-A. Venezuela. e-mail: apieters@ivic.ve

Susana El Souki. Biologist, UCV, Venezuela. Doctoral student, San Diego State University, USA.

César Nazar. Specialized Technician, Instituto Universitario de Tecnología de Yaracuy, Venezuela. Researcher assistant, Breeding Program for Sugarcane Improvement, INIA, Venezuela.

SUMMARY

A field experiment was set out to evaluate the differences in biomass production, nitrogen economy and grain production between two rice cultivars Araure 4 (A4) and Fonaiap 2000 (F2000), which differ in yield potential. Aboveground biomass and N distribution within the plant were determined at different growth stages and their relation to grain yield analyzed. A4 produced more biomass than F2000. However, yield and harvest index of A4 were 13 and 27% lower, respectively, than those of F2000. Spikelets per m² and the weight of 1000 grains were higher in F2000, but the percentage of filled grains was the same, resulting in a greater grain sink capacity in F2000. A4 maintained higher rates of biomass production than F2000 until anthesis; the opposite occurred during grain filling. Organ N concentrations were similar between cultivars, although leaves of A4 showed slightly larger values than leaves of F2000. A4 accumulated 15% more N than F2000 during the growth cycle. Most of the accumulated N (92%) in A4 was taken up between transplant and anthesis, whereas F2000 took up over 30% of total N between flowering and maturity, most of it for grain development. In F2000 60% of total N accumulated in the plant was in the grain whereas in A4 grains accounted for only 44%. Consequently, N harvest index and N use efficiency for grain production were much larger in F2000. The higher yield in F2000 compared to A4 can be partly attributed to greater sink capacity, larger rates of post-anthesis growth, N uptake and remobilization. These traits can compensate for a smaller plant size in F2000.

Crecimiento, rendimiento y economía de nitrógeno en dos cultivares de arroz bajo condiciones de campo en venezuela

RESUMEN

RESUMO

Estabeleceu-se um experimento de campo para avaliar as diferenças em produção de biomassa, economia de nitrogênio e produção de grão entre dois cultivares de arroz, Araure 4 (A4) e Fonaiap 2000 (F2000), con rendimentos potenciais diferentes. Determinaram-se a biomassa aérea e distribuição de N a diferentes estados de crescimento para relacioná-las com o rendimento. A4 produziu maior biomassa que F2000. No entanto, o rendimento e índice de colheita de A4 foram 13 e 27% menores, respectivamente, que os de F2000. As espiguinhas por m2 e o peso de 1000 grâos foram maiores em F2000, o que se traduziu em maior capacidade do sumidouro em F2000. A4 mostrou maiores taxas de produção de biomassa que F2000 até a antese; durante o enchimento do grão a tendência foi contrária. As concentrações de N nos órgãos foram similares entre cultivares. A4 acumulou 15% mais N que F2000 durante o ciclo de crescimento; 92% do N em A4 foi absorvido antes da antese, enquanto que em F2000 mais de 30% do N total se acumulou depois da floração, principalmente para desenvolvimento do grão. 60% do N total acumulado por F2000 foi para o grão, enquanto que em A4 esta fração foi 44%. Em consequência, o índice de colheita de N e a eficiência do uso de N em produção de grão foram maiores em F2000 que em A4. O maior rendimento de F2000, comparado com A4, pode atribuir-se parcialmente a uma maior capacidade do sumidouro, e maiores taxas de crescimento, acumulação e traslado de N para o grão. Estas características podem compensar a menor produção de biomassa em F2000.

KEYWORDS / Grain Production / Growth / Nitrogen Remobilization / Rice / Venezuela /

Introduction

In Venezuela, rice production has increased over ten-fold during the last 50 years due to a more than doubling of the cultivated land and to an over four-fold increase in yield over the same period (FAO, 2005). However, a significant stagnation of the inter-annual increase in yield has become evident during the last ten years compared to the period between 1960-1984, despite the development and release of a significant number of new cultivars between 1994 and 2005, which is comparable to the number of cultivars released and cultivated between 1960 and 1984 (Salih, 1996). This situation challenges the ability to meet the predicted increase in demand, estimated to be larger than the recent yield improvements.

Worldwide, several strategies have been proposed to increase rice yield potential as a means to increase rice production. These strategies range from the introduction of new management practices, the generation of new plant types, the introduction of foreign genes coding for enzymes involved in the C4 photosynthetic pathway and the use of hybrid rice. The advantage of the latter resides in its large capacity for translocation of pre-stored carbohydrates to the grain (Song et al., 1990).

Increases in rice yield can be achieved by either increasing biomass per unit of cultivated land, increasing the harvest index (HI; fraction of above ground biomass dedicated to grains) or both. Yield determinants are highly dependent on biomass production, N economy and distribution within the plant, and vary greatly depending on genotype, environment and management practices.

The HI is a complex trait that also depends, among other factors, on resource management, genotype, environment and, crucially, on the relation between sink capacity of the developing grain and the availability for grain development of either pre-stored or currently assimilated carbon by photosynthesis. However, it is not clear whether HI is source or sink limited, and how factors mentioned above modulate this limitation. In modern rice cultivars, the largest increase in yield potential came from increases in HI resulting from introduction of dwarfing genes (see Cassman, 1999), indicating that vegetative organs supply enough resources (C and N) to the grain and that yield could be a trait in fact sink limited. Evidence supporting this idea also arises from studies on transgenic rice with de-regulated ADP glucosa Pyrophosphorylase in the endosperm, which was found to produce more grain than untransformed plants (Smidansky et al., 2003). Nevertheless, there has been little improvement in HI in recent years, suggesting that HI is approaching a physiological ceiling (Peng et al., 1999). In addition, a high positive correlation has also been reported between biomass accumulation and yield in high-yielding cultivars released in the period 1985-1995 (Peng et al., 1999), which suggests that one alternative for increasing rice yield potential is by improving biomass production.

In monocarpic plants the onset of the reproductive stage is associated with senescence of vegetative organs. This involves nutrient remobilization (mainly N) from senescing tissues to the developing grains, which become the largest sink for C and N. Grain filling is also dependent on the metabolic competence of vegetative tissues, particularly of the flag leaf, in order to provide photo-assimilates to sustain grain growth. Between 60 and 90% of the C content in the panicle at the time of harvest is derived from photosynthesis of the flag leaf after heading (Yoshida, 1981). This situation presents a trade-off between nutrient remobilization to the grain during senescence and the maintenance of photosynthetic activity to provide C assimilates to reproductive organs (Mae, 1997).

In rice, remobilized N from vegetative organs accounts for 70-90% of panicle N (Mae and Ohira, 1981). Most of the remobilized N to the panicles comes from leaf blades and the rest is derived from leaf sheaths and culms (Mae, 1997). However, a large fraction of the accumulated N during the growth cycle remains in the straw at the time of harvest, ranging from 35% (Inthapanya et al., 2000) to over 50% of total N uptake (Ashraf and Hussain, 2005; Tao, 2004). Increasing the capacity to remobilize pre-stored N to the grain would represent a means to increase yield without the need for a heavier use of N fertilizer and/or breeding for plants with a larger potential for biomass production. Thus, analysis of the dynamics of biomass production and allocation, as well as patterns of N uptake and remobilization in tropical rice genotypes, is essential for understanding the possible mechanisms underpinning yield potentials achievable in these environments.

In this study two cultivars, Araure 4 (A4) and Fonaiap 2000 (F2000), which differ in their yield potential, were evaluated. Cultivar A4 is representative of the local germplasm cultivated during the '80s and early ‘90s. F2000 is a recently released cultivar with a yield potential ranging from 1.0 to 2.0tons·ha^-1, larger than that of A4. The aim was to quantify the differences in growth patterns, biomass production and allocation, and N use efficiency, in order to identify possible physiological and agronomical associations that might explain differences in yield between the cultivars. To achieve this goal, plants of F2000 and A4 were grown in the field and managed with the same agronomic practices of fertilization, plant density and pest management used by growers. The evidence suggests that a larger sink capacity and larger N remobilization and uptake during senescence associated with the reproductive stage can account for differences in grain production.

The rice cultivars A4 and F2000 were selected for this study. A4 was released in 1984, has a height from 100 to 110cm, a period of 120-135 days from germination to harvest and an average realized yield of 5.6tons·ha^-1 (Páez and Rodríguez, 1995). F2000 was released during the year 2000 by the National Institute of Agricultural Research (INIA, Venezuela). It shows a shorter height (80-100cm) than A4 and has a shorter period (110-120 days) from germination to harvest, with a predicted yield potential of up to 7.0tons·ha^-1. Both cultivars show good tolerance to most common pests and diseases and, due to sturdy stems and a relatively short height, are lodging resistant.

Seeds of F2000 and A4 were germinated in soil under glasshouse conditions. Twenty days later, on Jan 28 (A4) and Feb 16 (F2000) 2002, seedlings were transplanted manually to 20 plots (10 for each cultivar) of 5.0×4.0m, in the experimental fields of Fundación para la Investigación Agrícola DANAC, Yaracuy state, Venezuela, at 10º21'45''N and 68º39'0''W. Hill spacing was 0.3×0.25m with one plant per hill, for a final plant density of approximately 13 plants per m². Soils in the experimental plots were Gleyic Luvisol (FAO taxonomy). A water lamina of less than 10cm was maintained throughout the growth cycle until the final harvest. Fertilization was applied at a rate of 150kg·ha^-1 of N as urea, 50kg·ha^-1 of P and 70kg·ha^-1 of K. Application of P and K fertilizers was done as basal dressing 15 days after transplant (DAT) for each cultivar; N fertilisation was split into four applications at a rate of 60kg·ha-1 15DAT, and at 30kg·ha-1 on 30, 45 and 65DAT. Plants were harvested throughout the growth cycle starting 20DAT, at mid-tillering (MT), flowering (Fl) and maturity (M) stages. Plant material was taken to the laboratory and separated into organs (stems, leaf lamina and panicles), dried to constant weight in an oven at 65ºC and weighed. Leaf area per plant was measured immediately after organ separation with a leaf area meter (Licor, model 3100, Nebraska, USA) and used to calculate leaf area index (LAI).

Homogeneous sub-samples of the dried organs were ground in a mill and 50mg of the ground tissue was digested at 350ºC with concentrated H₂SO₄ using K₂SO₄ and CuSO₄ as catalyzers, and quantified by the microKjeldhal method. Specific leaf N (SLN) was calculated as N content in leaf biomass divided by the corresponding LAI.

Data were analyzed using a t-test following an F-test for homogeneity of variances (Sokal and Rohlf, 1969). When appropriate, comparisons were made for the same plant organ at the same growth stage on the two cultivars, or between cultivars within the same growth stage. Significant differences were assumed when P<0.05.

Cultivar A4 consistently produced more biomass than cultivar F2000 at all growth stages. The production of photosynthetic (leaf blades) and stem biomass was also larger in A4 than in F2000 (Figure 1). Differences in standing biomass between the two cultivars were mainly due to a larger leaf biomass at the flowering stage and to a much larger stem biomass at flowering in A4 than in F2000. Despite that plants of A4 grew larger than those of F2000, the latter produced more grain (13%) than A4. Consequently, HI in F2000 was significantly higher (Table I) than in A4. The number of spikelets per m² and the weight of 1000 grains were significantly larger in F2000 than in A4, whereas the percentage of filled grains, although smaller in A4 than in F2000, was not statistically significant (Table I).

A4 showed larger crop growth rates (CGR) than F2000 during vegetative growth (20DAT-Fl stages) but during the reproductive period (Fl-M) F2000 showed a CGR 13% larger than A4 (Figure 2). Although differences in biomass production and CGR were found between both cultivars, LAI was significantly larger (P<0.05) in A4 than in F2000 (Figure 3) only 20DAT. Thereafter, no significant differences were observed until the M stage, when F2000 had a significantly larger LAI than A4 (P<0.05).

Leaf and stem N concentrations decreased with age in both cultivars (Figure 4). Leaf N was similar between the two cultivars until the reproductive stage (Fl), when N concentration in leaves of A4 was larger than in F2000. The same trend was observed in stems at the MT stage. The N remaining in dead biomass (leaf lamina and sheath) was similar between the two cultivars during senescence (M stage). However, N concentration in stems of A4 was significantly larger than in F2000 at maturity. Grains showed no significant differences in N concentration between cultivars (P>0.05; Figure 4).

Consistent with the highest leaf N concentration found in both cultivars 20DAT, the specific leaf nitrogen (SLN) was also greatest at this growth stage (Figure 5). Although a slight increase was observed in A4 at the M stage, no statistically significant differences were observed between cultivars.

Twenty DAT, N content of A4 was over 2-fold larger (P<0.05) than that of F2000 (Figure 6). At maturity the N content in stems of A4 was larger than in F2000. The opposite was observed for grains: F2000 had a grain N content 25% larger than A4 (Figure 6). At final harvest, stems of A4 showed N content 2-fold larger than those of F2000, whereas grains of F2000 had 15.8% more N accumulated than grains of A4. Accordingly, plants of A4 took up ca. 15% more N than F2000 (Figure 6) and this trend was maintained throughout the growth cycle. Differences in N uptake between genotypes ranged from 20 to 60% more N accumulated in A4 than in F2000, depending on the growth stage. Rates of N uptake were highest during the period between 20DAT and MT in both cultivars (Figure 7). However, patterns of N uptake during the growth cycle differed between cultivars. Whereas in A4 N uptake progressively declined after the MT stage, the lowest rate of N uptake in F2000 occurred between MT and Fl stages, to resume N absorption during the grain filling period (Fl-M), compared to the period between MT and Fl stages.

The results presented showed important differences in growth, yield, N uptake and distribution between two tropical rice cultivars, A4 and F2000. A4 grew larger, producing over 25% more biomass than F2000. This difference was mainly due to a larger leaf biomass produced during vegetative growth and a larger stem biomass produced between mid-tilling and flowering in A4, which accounted for almost 30% of total above ground biomass in this cultivar.

The smaller size of F2000 at harvest can be partly attributed to its shorter growth period (20 days shorter) compared to A4. Akita (1989) reported a linear positive relationship between growth duration and total biomass production in rice. Dry matter accumulated before anthesis has been reported to contribute importantly to grain yield in cereals (Osaki et al., 1991; Pheloung and Siddique, 1991). However, it has been recently shown that final grain yield in rice is also closely associated with post-anthesis growth (Takai et al., 2006). Consistent with reports by Takai et al. (2006) we found that the rate of biomass production between flowering and maturity was 13% larger in F2000 than in A4 (see Figure 2), which was associated with a larger grain yield and HI than A4.

The higher post-anthesis biomass accumulation in F2000 coincided with the resumption of N uptake, whereas in A4 post-anthesis N uptake was the lowest during its growth cycle. It has been suggested that the amount of N accumulated during the reproductive phase determines sink size (Hasegawa et al., 1994), as N absorbed during this period is mostly dedicated to reproductive organs rather than to photosynthetic tissues. In fact, F2000 produced larger grains and more spikelets per m² than A4, indicating a higher sink capacity in the former. The coupling of these two processes (larger biomass production and N uptake at maturity) in F2000 seems to be crucial in determining the grain sink capacity and therefore, the higher yield shown by this cultivar.

No differences were found in SLN (N per unit of leaf area) between A4 and F2000, except during the maturation stage. Given the close relation between N per unit leaf area and photosynthetic potential (Peng et al., 1995), the present results suggest no differences in photosynthesis on a leaf area basis between A4 and F2000. Pieters and El Souki (2005), using chlorophyll fluorescence techniques, reported a similar photosynthetic capacity in the flag leaf of A4 and F2000 grown under glasshouse conditions. The higher N concentration in leaves of A4 at maturity could explain the partial recovery of the SLN observed in this cultivar at this stage (see Figures 3 and 4). Nevertheless, the higher leaf N concentration at maturity in A4 seems to have contributed little to grain production in this cultivar. Rather, this higher N concentration is the consequence of a slower rate of N remobilization in this cultivar. Ying et al. (1998b) analyzed a number of rice cultivars in tropical and sub-tropical environments and found little differences in SLN despite large environment-dependent differences in yield (Ying et al., 1998a), suggesting that photosynthetic rate on a leaf area basis may not be a direct determinant of grain production as has been reported earlier (Murchie et al., 2002). These authors suggest that remobilization of C reserves from leaf sheaths and culms may have the effect of de-coupling photosynthesis from grain filling (Murchie et al., 2002). However, the cultivars analyzed in the present study differ not only in size, but also in plant architecture. F2000 is an erect cultivar and culms of A4 show a more open habit. Duncan (1971) suggested that an erect posture of upper leaves within a rice canopy provides a better light distribution within the canopy, allowing lower leaves to contribute to a larger extent to the daily carbon gain of the plant due to a higher radiation use efficiency (see also Long et al., 2006). The advantage of an erect posture over an open one is greatest in clear days (Ort and Long, 2003), as was the case during this experiment (dry season). Thus, plants of F2000 could have had a greater photosynthetic capacity per plant and hence a larger C gain than A4. This trait together with a larger N and probably C remobilization from leaves and stems during grain filling could, in part, explain the larger yield of F2000, as has been suggested for hybrid rice (Song et al., 1990). This possibility requires further experimental testing.

N concentration in leaves of F2000 was lower than in A4 at flowering and at maturity, but N concentration of dead biomass was the same in both cultivars and stems of A4 had more N than those of F2000. These results indicate a preferential and more efficient N translocation in F2000 than in A4 during senescence, and suggest that a higher remobilization capacity can compensate for, and overcome, a smaller biomass production and a lower N accumulation during vegetative growth, emphasizing the important role of N remobilization in grain yield. Notably, F2000 resumed N uptake during grain filling compared to A4, which showed the lowest rate of N uptake at this stage (see Figure 6). It is evident that the extra N absorbed by F2000 during grain filling was not involved in maintaining photosynthesis as SLN and leaf N concentration decreased during this period. It is more likely that the larger N uptake in F2000 between flowering and maturity was associated with the larger growth rate observed during this period and was most likely invested in the production of grain.

A larger capacity for translocation of pre-stored N to the grain in F2000 seems to be crucial in explaining the differences in grain yield reported here and is able to compensate for and overcome a shorter growth duration and smaller production of vegetative biomass.

A faster post-anthesis crop growth and maintenance of N absorption during grain filling in F2000 seem to be mechanisms associated with the larger yield observed in this cultivar.

The authors thank Ernesto Medina for his comments on an earlier version of the manuscript, and Carla Alceste at IVIC and Oranys Marín at UCV, for nitrogen determinations. This research was funded by Instituto Venezolano de Investigaciones Científicas (IVIC) and Fondo Nacional de Ciencia y Tecnología (FONACIT) grant S1-2001000904. IVIC receives financial support from the Ministry of Science and Technology of Venezuela.

1. Akita S (1989) Improving yield potential in tropical rice. In Progress in irrigated rice research. International Rice Research Institute. Los Baños, Philippines. pp 41-73        [ Links ]

2. Ashraf M, Hussain F (2005) Dry matter and nitrogen distribution at maturity of three rice (Oryza sativa L.) cultivars exposed to ammonia at two growth stages. J. Agron. Crop Sci. 191: 125-129.        [ Links ]

3. Cassman KG (1999) Ecological intensification of cereal production systems: yield potential, soil quality, and precision agriculture. Proc. Nat. Acad. Sci. USA. 96: 5952-5959.        [ Links ]

4. Cassman KG, Harwood RR (1995) The nature of agricultural systems: food security and environmental balance. Food Policy 20: 439-454.        [ Links ]

5. Duncan WG (1971) Leaf angle, leaf area and canopy photosynthesis. Crop Sci. 11: 482-485.        [ Links ]

6. FAO (2005) FAOSTAT database. http://faostat.fao.org/faostat        [ Links ]

7. Fillery IRP, Simpson JR, De Datta SK (1986) Contribution of ammonia volatilization to total nitrogen loss after application of urea to wetland rice fields. Fert. Res. 8: 193-202.         [ Links ]

8. Fisher KS (1998) Toward increasing nutrient-use efficiency in rice cropping systems: The next generation of technology. Field Crop Res. 56: 1-6.        [ Links ]

9. Hasegawa T, Koroda Y, Seligman NG, Horie T (1994) Response of spikelet number to plant nitrogen concentration and dry weight in paddy rice. Agron. J. 86: 673-676.        [ Links ]

10. Inthapanya P, Sipaseuth P, Shihavong V, Sihathep M, Changphengsay S, Fukay S, Basnayake J (2000) Genotype differences in nutrient uptake and utilization for grain yield production of rain-fed lowland rice under fertilized and non-fertilized conditions. Field Crop Res. 65: 57-68.        [ Links ]

12. Long SP, Zhu X, Naidu SL, Ort DR (2006) Can improvement in photosynthesis increase crop yields? Plant Cell Envir. 29: 315-330.        [ Links ]

13. Mae T (1997) Physiological nitrogen efficiency in rice: Nitrogen utilization, photosynthesis, and yield potential. Plant Soil 196: 201-210.        [ Links ]

14. Mae T, Ohira K (1981) The remobilization of nitrogen related to leaf growth and senescence in rice plants (Oryza sativa L.). Plant Cell Physiol. 22: 1067-1074.        [ Links ]

15. Murchie EH, Yang J, Hubbart S, Horton P, Peng S (2002) Are there any associations between grain-filling rate and photosynthesis in the flag leaves of field grown rice? J. Exp. Bot. 53: 2217-2224.        [ Links ]

16. Ort DR, Long SP (2003) Converting solar energy into crop production. In Chrispeels MJ, Sadava DE (Eds.) Converting solar energy into crop production. American Society of Plant Biologists/Jones and Bartlett. Boston, MA, USA. pp 240-269.        [ Links ]

17. Osaki M, Sinano T, Tadano T (1991) Redistribution of carbon and nitrogen compounds from the shoot to the harvesting organs during maturation in field crops. Plant Nutr. 37: 117-128.        [ Links ]

18. Páez O, Rodríguez R (1995) Evaluación del rendimiento de los cultivares de arroz Araure 3 y Araure 4 bajo tres densidades de siembra. Agron. Trop. 54: 283-292.        [ Links ]

19. Peng S, Cassman KG, Kropff MJ (1995) Relation between leaf photosynthesis and nitrogen content of field grown rice in the tropics. Crop Sci. 35: 1627-1630.        [ Links ]

20. Peng S, Cassman KG, Virmani SS, Sheehy J, Khush GS (1999) Yield potential trends of tropical rice since the release of IR8 and the challenge of increasing rice yield potential. Crop Sci. 39: 1552-1559.        [ Links ]

21. Pheloung PC, Siddique KHM (1991) Contribution of stem dry matter to grain yield in wheat cultivars. Aust. J. Plant Phys. 18: 53-56.        [ Links ]

22. Pieters AJ and El Souki S (2005) Effects of drought during grain filling on PS II activity in rice. J. Plant Phys. 162: 903-911.        [ Links ]

23. Salih LA (1996) Mejoramiento genético de arroz en Venezuela. In INGER (Ed.) INGER América Latina - Informe 1995. Red Internacional para la Evaluación Genética del Arroz. INGER-América Latina. Cali, Colombia. pp. 11-17.        [ Links ]

24. Smidansky EM, Martin JM, Hannah LC, Fischer AM, Giroux MJ (2003) Seed yield and plant biomass increases in rice are conferred by de-regulation of endosperm ADP-glucose pyrophosphorylase. Planta 216: 656-664.        [ Links ]

25. Sokal RR, Rohlf FJ (1969) Biometry. Freeman. San Francisco, CA, USA. 776 pp.        [ Links ]

26. Song X, Agata W, Kawamitsu Y (1990) Studies on dry matter and grain production of F1 hybrid rice in China. II. Characteristics of grain production. Jap. J. Crop. Sci. 59: 29-33.        [ Links ]

27. Takai T, Matsuura S, Nishio T, Ohsumi A, Shiraiwa T, Horie T (2006) Rice yield potential is closely related to crop growth rate during late reproductive period. Field Crop Res. 96: 328-335.        [ Links ]

28. Tao H (2004) Yield and nitrogen uptake of lowland rice (Oryza sativa L.) in a water saving ground cover rice production system (GCRPS) in Beijing, North China. Thesis. Christian-Albrecht University of Kiel, Germany. 256 pp.        [ Links ]

29. Xie JC (1998) Present situation and prospects for fertilizer use. Plant Nut Fert Sci. 4: 321-330.        [ Links ]

30. Ying J, Peng S, He Q, Yang H, Yang C, Visperas RM, Cassman KG (1998a) Comparison of high-yield rice in tropical and subtropical environments. I. Determinants of grain and dry matter yields. Field Crop Res. 57: 71-84.        [ Links ]

31. Ying J, Peng S, Yang G, Zhou N, Visperas RM, Cassman KG (1998b) Comparison of high-yield rice in tropical and subtropical environments. II. Nitrogen accumulation and utilization efficiency. Field Crop Res. 57: 85-93.        [ Links ]

32. Yoshida S (1981) Physiological analysis of rice yield. In: Fundamentals of rice crop science. International Rice Research Institute. Makita, Philippines. pp. 231-251.        [ Links ] ]]> 1 1989 41-73 2 2005 191 125-129 3 1999 96 5952-5959 4 1995 20 439-454 5 1971 11 482-485 6 FAO 2005 7 1986 8 193-202 8 1998 56 1-6 9 1994 86 673-676 10 2000 65 57-68 12 2006 29 315-330 13 1997 196 201-210 14 1981 22 1067-1074 15 2002 53 2217-2224 16 2003 240-269 17 1991 37 117-128 18 1995 54 283-292 19 1995 35 1627-1630 20 1999 39 1552-1559 21 1991 18 53-56 22 2005 162 903-911 23 1996 11-17 24 2003 216 656-664 25 1969 776 26 1990 59 29-33 27 2006 96 328-335 28 2004 256 29 1998 4 321-330 30 1998 a 57 71-84 31 1998 b 57 85-93 32 1981 231-251.